Turn any video into a Mixamo-rig animation - GVHMR estimator, spec-driven retarget, FK apply in Blender via MCP. Works with any Mixamo character; built to be operated end-to-end by an AI agent.
-
Updated
Aug 18, 2026 - Python
#
gvhmr
Here are 6 public repositories matching this topic...
Turn locked-camera video into clean FK animations on Mixamo characters—no suit, no keyframing, fully scriptable for AI agents.
deep-learning mmd mocap bvh attention-mechanism agents from-scratch 3dsmax retargeting motionbuilder finetuning vtuber romp large-language-models ggml gvhmr
-
Updated
Aug 26, 2026 - Python
Turn any video into controllable 3D avatar motion: extract SMPL, retarget to a VRM avatar, let an LLM understand the movement, and learn it with a realtime AI voice coach. Open-source, deterministic motion pipeline.
threejs avatar computer-vision animation gemini motion-capture dance vrm pose-estimation motion-retargeting smpl fastapi llm ai-coach gvhmr
-
Updated
Aug 17, 2026 - Python
I know kung fu. — A Construct-style training dojo for traditional kungfu (qigong, taichi, etc.), powered by humanoid simulation. SMPL-X + Unitree G1 dual-track playback with multi-view rendering.
humanoid-robot motion-retargeting mujoco gmr viser smpl-x qigong baduanjin unitree-g1 gvhmr genesis-simulator kung-fu human-motion-recovery kinematic-playback roboharness
-
Updated
May 25, 2026 - Python
Improve this page
Add a description, image, and links to the gvhmr topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the gvhmr topic, visit your repo's landing page and select "manage topics."